app.py

import numpy as np
import pandas as pd
import streamlit as st

from utils import get_index, get_movie_info

MAX_OVERVIEW_LENGTH = 280
IMAGE_WIDTH = 170


@st.cache()
def init() -> None:
    movies = pd.read_csv("movies.csv")
    st.session_state["movies"] = movies
    st.session_state["titles"] = movies["original_title"].values
    st.session_state["sim_matrix"] = np.load("sim.npy")


def run() -> None:
    st.title("Movie Recommender")
    with st.form("search"):
        title = st.selectbox("Recommend movies based on:", st.session_state["titles"])
        num_recommendations = st.number_input(
            "Number of recommendations:", min_value=1, max_value=20, value=6
        )
        search = st.form_submit_button("Search")
    if search:
        recommend(title, num_recommendations)


def recommend(based_on: str, num_recommendations: int) -> None:
    movie_index = get_index(based_on)
    similarities = list(enumerate(st.session_state["sim_matrix"][movie_index]))
    similarities.pop(movie_index)
    sorted_similarities = sorted(similarities, key=lambda x: x[1], reverse=True)
    recommendations = sorted_similarities[:num_recommendations]
    st.markdown("---")
    for idx, _ in recommendations:
        show_movie_info(idx)


def show_movie_info(movie_index: str) -> None:
    info = get_movie_info(movie_index)
    im_col, info_col = st.columns([2, 4])
    im_col.image(
        f"https://a.ltrbxd.com/resized/{info['image_url']}.jpg", width=IMAGE_WIDTH
    )
    info_col.markdown(
        f"### {info['original_title']} ({info['release_date'].split('-')[0]})"
    )
    info_col.markdown(f"*{int(info['runtime'])} min*")
    short_overview = (
        f"{info['overview'][:MAX_OVERVIEW_LENGTH]} ..."
        if len(info["overview"]) > MAX_OVERVIEW_LENGTH
        else info["overview"]
    )
    info_col.markdown(short_overview)
    info_col.caption(", ".join(eval(info["genres"])))
    st.markdown("---")


if __name__ == "__main__":
    init()
    run()